Summer only postponed

On Friday, 20th July at 16:00 hrs (in the middle of the English Summer) we set of for home in Worcestershire from our week-day-job in Newport in Gwent, Wales. The journey was made unpleasant by the non-stop rain. Half an hour later we were driving through flood water 60cm (2 ft.) deep. Cars were spluttering to a halt, and then marooned, as total chaos hit the Counties of Worcestershire and neighbouring Gloucestershire.

With great skill, Margaret managed to get us of the Motorway, and phone enquiries revealed that the flash floods had made it impossible to reach home. So we settled down for a night in the car in the country, in the company of several hundred other vehicles, while the non-stop rain continued to pour down its misery. The flash-floods flowed into the rivers and the Severn and Avon breached their banks. We managed to reach home the next day, the whole journey normally takes 1 and a half hours, this time it took 22 hours!

A week later, after a few days respite from rain, how did the Cannas take it? Well they are tough old plants, but there is hardly any decent foliage, as well as heavy shredding, many leaves have yellow stripes or mosaic, and we have had to help many plants by unravelling the leaves manually. At any other time, I would have considered that they were suffering with Canna virus, but after two months of hardly any direct light and only a handful of days without cold rain, capped by a months worth of rain hitting them in 2 hours, I am amazed at how strong they still are and logically I must attribute the problems to plant stress.

The technical reasons for the poor state of the foliage is due to several factors:

• Oxygen is needed in the soil to help roots with uptake processes. Where there is no oxygen, such as in flooded sites, sugar cannot be utilised by the plants to produce energy for nutrient uptake.
• Nutrient uptake is faster in warmer soils than in cold soils.
• Plants under stress are less able to take up nutrients, generally due to a reduced or damaged root system.

Comparing the Canna beds with the vegetable plot; the potato main crop has been flattened after potato blight hit, and many tubers are rotten as I dig them up, most of the onions have gone rotten as they grew in the ground, beans are half their normal height, and salad crops have been physically flattened. Basically... everything destroyed. On that basis I can only wonder at the resilience of the Cannas.

So, with a week free of rain promised by the weather man, we now need to help the Cannas recover. So, they have each already had a handful of Vitax Q4 pellets, an all-purpose fertiliser which contains both vital nutrients and trace elements essential for plant growth. This is almost double the price of the Growmore feed that we normally use, but the latter does not contain trace elements and we are working on the basis that the water inundation for two months has leached or washed away much of the nutrients and trace elements. The Q4 also has a higher ratio of potash, which is required by flowering plants, and will be taken up over the next month or so as the pellets dissolve, and therefore this is a long-term solution.

After a few days of sunshine, and the gradual drying out of the soil in the beds and the peat in the pots, the Cannas they will also get a liquid feed of Miracle-Gro to give them an immediate kick-start. Both growing mediums are sodden at present, and anymore liquid will be detrimental. The liquid feed will be repeated after a further week. This is our short-term solution.

Next weekend, the Cannas will get a severe grooming, totally removing stalks with nothing to recommend their foliage, and the growth rate of the new stalks should be fast without the old stalks to compete for resources. In the meantime, we have to hope that summer has only been postponed and not cancelled!

Systemic sprays give virus protection

Just read an interesting paper, Studies on the relationship of canna mottle virus and its vector myzus persicae, authored by M. Datta-Gupta & S.P. Raychaudhuri in 1975, published in the Indian Journal of Entomology 37(4): pages 336-343. So, this again proves that Canna virus is not a recent phenomenon, that paper was over 30 years old. Moving on, myzuz persicae is the formal name for the common-a-garden aphid, the culprit that spreads Canna Mottle Virus.

Interestingly, the paper stated that when a contaminated aphid eats at the Canna leaf for 1 minute it is contaminated. Equally, an uncontaminated aphid eats the foliage of a contaminated Canna, then it is also contaminated in one minute.

This started me thinking; once an aphid starts eating the leaf of a Canna protected by a systemic insect spray it dies within seconds. On that basis we can assume that regular spraying with a systemic insect spray can protect our Cannas from the virus.

This leaves us with a dilemma, those of us trying to advocate non-chemical solutions cannot win this particular argument, because it may be an hour before a ladybird ascends on the aphid, and the damage is done.

This brings us to ants. A drop of honey, or something like it, is the connecting bond between the ant and the aphids. It is exuded by the latter through certain tubercles which are situated at the end of the abdomen, and is, of course, the product of the endless quantities of sap, which, so long as it lasts, these insects are for ever pumping up and swallowing from the plant they inhabit. This honey, or honey-dew, to use the more special name bestowed on it, the ants want, but they are not content with drinking it whenever it issues from its manufacturers, in natural course. This is not sufficient, and they have learned to increase the flow of so valued a beverage by their own efforts, in other words, they milk the aphids, which thus become their cows. To do this they tap them with their antennæ, softly and gently, on the sides of the abdomen - a quick little shower of touches. Under the influence of this probably pleasant sensation the aphids becomes willing to part, and, raising the abdomen, "teems her refreshing dew" in a drop from the tip of it. This action of the ants cannot, in Europe, be successfully imitated, at least it has not been, and if an ant is not forthcoming the fluid is contained in the body of the aphids until necessity compels its being ejected. Probably the ants, if delayed in their visits, are missed by the aphids, as a cow misses her milker, and long before they do excrete, as the process is called, they would perhaps have done so had they felt able. The sensation no doubt of the ant's antenne on the abdomen has become, through usage, the almost necessary stimulus to the act produced by it.

So, logically, we also have to destroy ants, as they are the next step in the vector. They farm the aphids and protect them from predators, and if they are also eliminated then that reduces the risk of virus contamination considerably, by reducing aphid activity considerably. However, there is one more aphid fact that I picked up recently. Flying aphids can travel up to 10 miles, using the winds to keep them afloat for considerable periods. So, if anyone within 10 miles has virus on their Cannas and the wind is in the right direction, then we are all vulnerable, no matter how we try to protect our plants. However, if our plants have a systemic protection, then they will kill the aphid before it contaminates the plant.

Torrential rain again at Worcester

Torrential rain is sweeping across the UK, with flash floods leaving some homes waterlogged and schools closed. The band of severe weather has hit southern and central England and Wales, and will move north to reach Scotland and Northern Ireland over the weekend. Rain is falling at up to 18mm (0.7in) an hour and some areas could see levels reach 100mm (3.9in) in 24 hours. The Environment Agency has two flood warnings and the Met Office has severe weather warnings for England and Wales.

In the meantime, at Worcester, our Cannas, both the mature plants and this years seedlings are taking a beating. The collection looks so sad, there are no more than 50% of the plants showing good foliage. The lack of direct sunshine for nearly two months is having a dire effect. We have had only a handful of days with direct sunlight, when it has not been raining it has been overcast with low cloud.

We have to help many plants to unfurl their leaves, the lack of light means a lack of energy. Photosynthesis is the conversion of light, and is the reason d’etre for Cannas fabulous foliage. More plants have flowered this month than in previous years, and they are also shorter than usual. Obviously, this weather is creating a great deal of stress, for the Cannas and us, and this weekend they are going to have to endure another wall of water hitting them for hours on end, having barely survived similar torrents last month. The seedlings are not yet advanced enough to be able to send up another shoot quickly if the main stem is damaged, and we will be looking this weekend at creating more cover for the worst affected.

New Italian cultivars

An article by Professor F. A. Waugh, University of Vermont, taken from the Garden and Forest Journal [Volume 10, Issue 501. [September 29, 1897, 379-388]

Notes on the Orchid-flowering Cannas.

WE have been greatly interested in the new Orchid-flowering Cannas since their first announcement, and have taken some pains to grow most of the varieties offered for sale. A note regarding the variety America was sent to GARDEN AND FOREST some weeks ago (see vol. x., page I78), and a completer account has been more recently published in the Tenth Annual Report of the Vermont Experiment Station, page 119. Since that report went to press we have brought two new varieties of these Cannas into blossom. These are Bavaria and Burgundia, both from the lists of Dammann & Co., the originators of Italia, Austria and America. We are somewhat disappointed in these two newest varieties, not because of bad qualities, but on account of their close resemblance to Italia. Italia is a beautiful Canna, to be sure, perhaps the best one of this class yet introduced, but Burgundia and Bavaria are so much like it that a careless observer would pass them by as all of the same kind.

Both varieties are, however, of smaller stature than Italia and have smaller foliage. The flowers in all three are of a brilliant canary-yellow upon which two shades of rich apricot-red are successively overlaid. In Italia the red colors are run together in the throat to make somewhat regular solid blotches bordered with very deep bands of the clear yellow, like an exaggerated Queen Charlotte. In Bavaria there is very little of the darker red shade, while the lighter red is scattered in small dots well out upon the petal-like staminodia, giving an effect more like that of Florence Vaughan. Burgundia is almost a medium between Bavaria and Italia, with more of the dark overlying red than the former, and more of the leopard spottings than in the latter. Burbank, the only variety of American origin yet advertised, does not seem to have been noted so much in the horticultural press as its good qualities deserve. It is the equal of the Italian varieties. The flowers have the form of Italia and nearly the coloring of Austria. That is to say their form is the best, and their coloring pure canary yellow with a few faint reddish spots in the throat.

These new Cannas, hybrids of Madame Crozy with C. flaccida, are especially admirable for two qualities, namely, the perfection of form and the richness of color in their blossoms. The type of Canna-flower which we are used to seeing among the French dwarfs, when taken by itself, is singularly inharmonious and unsatisfying in its outlines, but there is a fullness of form and grace of outline among the Orchid-flowering sorts which by comparison is altogether pleasing. Their richness of coloring is remarkable. The comparative size of the flowers has been considerably overstated in the advertisements. They are quite large enough, to be sure, but they are only a little larger in fact than Charles Henderson, Alphonse Bouvier, or dozens of other old and well-known varieties. Several of these varieties we have had this year in quantities sufficient for liberal use in outdoor beds. Their large, luxuriant, Musa-like foliage is quite effective, but there is a noticeable paucity of flowers in comparison with the older French dwarf sorts. The softness and flaccidity of the blossoms detract somewhat, of course, from their usefulness out-of-doors, but not so seriously as we had expected. The flower-spikes, though comparatively few in number, are fairly durable and effective as far as they go.

The Orchid-flowering Cannas-and, by the way, they ought to have a better class name-are certainly attractive novelties for the amateur, but in their present state they are not likely to find great favor among professional gardeners, who are interested chiefly in gaudy red and yellow floral effects.
University of Vermont. F. A. Waugh.

Sprenger was a busy botanist

It is well known that the Italian Group of Cannas were first introduced by Carl Sprenger of Dammann and Co., closely followed by Luther Burbank and then taken up in a big way by Antoine Wintzer, Vilmorin-Andrieux, Wilhelm Pfitzer and others. However, what is not so well known is the extent to which Sprenger introduced new varieties.

The book Das Geschlecht der Canna by Árpád Mühle, 1909, provides a list of all of his Italian Group cultivars.

1. C. 'Africa'

2. C. 'Alemania'

3. C. 'America'

4. C. 'Aphrodite'

5. C. 'Asia'

6. C. 'Atlanta'

7. C. 'Australia'

8. C. 'Austria'

9. C. 'Bavaria'

10. C. 'Borussia'

11. C. 'Brittania'

12. C. 'Burgundia'

13. C. 'Campania'

14. C. 'Ch. Naudin'

15. C. 'Edouard André'

16. C. 'Emilia'

17. C. 'Grand Sasso d'Italia'

18. C. 'H. Wendland'

19. C. 'Heinrich Seidel'

20. C. 'Hellas'

21. C. 'Iberia'

22. C. 'Italia'

23. C. 'Kronos'

24. C. 'La France'

25. C. 'Oceanus'

26. C. 'Pandora'

27. C. 'Partenope'

28. C. 'Penelope'

29. C. 'Perseus'

30. C. 'Phoebe'

31. C. 'Pluto'

32. C. 'Prof. Treub'

33. C. 'Rhea'

34. C. 'Roi Humbert'

35. C. 'Roi Leopold'

36. C. 'Roma'

37. C. 'Romagna'

38. C. 'Sicilia'

39. C. 'Suevia'

40. C. 'Trinacria'

41. C. 'Umbria'

42. C. 'Wilhelm Beck'

What is not so well know, is that, Sprenger also released many new Crozy Group cultivars prior to exciting the world with the release of C. 'Italia' and C. 'Austria' in 1893. Some of those are:

1. C. 'Aetna'
   
2. C. 'Ajax'
3. C. 'Alkmene'
4. C. 'Arethusa'
5. C. 'Capri'
6. C. 'Königin von Italien'
7. C. 'Kronprinz von Italien'
8. C. 'Madame Oscar Meuricoffre'
9. C. 'Nero'
10. C. 'Paestum'
11. C. 'Prinzessin Laetitia'
12. C. 'Professor Rossi'
13. C. 'Solfatara'
14. C. 'Stromboli'
15. C. 'Sultana'

This information was first published in 1909, and as Dammans' nursery was destroyed by the eruption of Mount Vesuvius in 1906, then this is likely to be the most complete list of Sprengers' introductions that we are going to be able to collate.

Canna virus

Like all animals and plants, the Canna genus is susceptible to certain viruses, which may result in spotted or streaked leaves, in a mild form, but can finally result in stunted growth and twisted and distorted blooms and foliage. Known species of virus, confirmed by Dr Rick Mumford, senior virologist at the Central Science Laboratories in England, are:

1. Canna yellow mottle badnavirus (CYMV) infecting canna species.
2. Bean yellow mosaic virus (BYMV) infecting cannas, gladiolus, freesia and many legumes.
3. Tomato aspermy virus (TAV), causes mosaic in cannas, but it has not been reported affecting cannas in the UK.
4. Cucumber mosaic virus (CMV), cannas are susceptible to this virus, but none found yet in England.
5. Canna yellow streak virus (CaYSV), this was recently discovered by scientists at the Central Science Laboratory in England. Dr Rick Mumford is quoted as stating “Typical virus symptoms include flecking, mosaic, leaf streaking and necrosis, which in severe cases render plants unsaleable.”
   

To put this in context, humans have dozens of virus to contend with, and the Dahlia enthusiasts have to live with over twenty different virus's, but they have learnt to live with them. Whenever an affected specimen is spotted, then it is burnt. In addition, the commercial growers do not attempt to sell affected stock. The latter is still not happening in the Canna world, and we are still seeing many commercial growers selling their badly affected stock. Once the commercial growers stop spreading this viral pollution, the genus will stabilise and the virus will stop being a major topic of Canna conversation.

Overall, very little is known about the Canna viruses, but the following points are generally accepted:

• It manifests itself in rust coloured streaks or mottled markings on the leaves and in colour breaks on the flowers.
• Sometimes leaves are slightly distorted and puckered.
• Like many plants under stress, affected cannas will flower very early in the season and before the plant is full height.
• Over the years the canna will loose vigour and become increasingly unsightly.
• Some are spread by aphids and other sap sucking insects.
• At one time it was thought that Cannas may have the ability to outgrow the virus, this has not been proven. It may be that stressed plants simply recover, and as the stressed plant exhibits similar leaf damage to virus it was assumed wrongly that the plant had virus.
  
• Most authorities advise to burn all affected cannas and start again.
• Keep any new introductions well away from potentially infected stock.

The point about plant stress being mistaken for virus is important, and I would recommend that you do not destroy plants until you are sure that they have virus. I quarantine any suspects, so that they cannot pass on virus, then they are spoilt rotten. Best soil, best sun aspect, best protection against hard weather, best food, best watering possible.

After a month, or so, of this life of luxury, if they are not looking healthy, then they are presumed to have virus and are destroyed.

Grooming Cannas

Dead-heading cannas assures continuous blooms throughout their extraordinarily long flowering season. Do not allow dead flowers to accumulate on the ground, as this only provides for unhealthy growing conditions. Instead, collect and put on the compost heap. If lower leaves look weather worn then cut off and compost. It will help to open-up the air circulation around the plant and in a wet summer that will be very beneficial.

When all of the flowers on a branch have been exhausted then remove the branch on which on which they were borne. Normally another branch on the same stem will already bear opening flowers. Continue removing branches when all their flowers are spent, until the stem stops growing new branches.

You can see in the photograph how all of the flowers have been deadheaded from the branch on the right and there are only seed pods left. The branch on the left has many new buds getting ready to open. At the point marked with "Next" is where the next branch will emerge when the new branch has produced most of its flowers. Most cultivars will produce 3-5 branches before the stem becomes exhausted.

If the exhausted branches are not removed then the plant will waste valuable nutrition growing seed pods and the extra flowering branches will not develop. However, eventually the stem will stop throwing branches, at which time the whole stem should be cut off just above ground level so that the new emerging stems do not have to compete for light and nutrients and also improve the air flow around the lower stems. In addition, the foliage on the old stem will start to look jaded and the appearance of the whole plant is improved by the stem pruning.

Mulch to save moisture and weeding

If you want to spend your summer relaxing in the garden, rather than weeding and watering, try mulching beds and borders – a thick layer of leaf mould, composted bark or other covering material, will make maintaining borders easy and give your Cannas and other plants a boost at the same time.

Why mulch?
There are many reasons why mulching the soil every year is a good idea, the main one being that a garden hoe is a dangerous weapon in a bed of Cannas; the fragile young shoots get decapitated with monotonous regularity!. Deep mulch will prevent weeds from growing, which rob moisture from the soil and need constant hoeing or pulling out by hand to control. Although some weed seeds will germinate on top of the mulch, they cannot anchor themselves well in the bulky material and are easy to uproot.

Although winter rain can wash goodness out of the soil, you can give it a pick-me-up by spreading mulch - as the material rots down nutrients are released back into the earth. If applied over wet ground in the spring, it will also help to lock in moisture that can be used by plants in warmer weather. This is a real boon given recent water shortages and the constant threat of hosepipe bans.

Mulches can also be spread over Cannas or around the stems of less hardy plants in winter to help insulate them from frost.

Which mulch?
If you have made your own leaf mould or garden compost, this is ideal, but don’t worry if you haven’t. Garden centres and DIY stores stock a large range of materials that can be used instead, including well-rotted farmyard manure, composted bark and organic wheat straw.

Choose mulch that will suit your garden - wheat straw sets off Cannas and other tropical plants wonderfully, but looks out of place in a traditional border. Probably the most popular material is bark, which is available in several grades. As a rule of thumb: the finer the grade, the quicker it will rot down and need replacing.

In some gardens gravel or slate shingle makes an attractive mulch – lay it on top of a sheet of landscaping fabric (available from garden centres) to prevent it from being trodden into the soil. In addition, a stone mulch has the property of retaining ground heat overnight. That is an advantage in a temperate climate, but may be disadvantageous in a sub-tropical climate.

How to use?
Before laying a mulch, clear the site of weeds and make sure the soil is moist. If necessary soak well as it will be difficult to wet the soil after it has been laid. Spread a 10cm (4in) layer of material across the whole area with a spade and rake to leave a level finish - wear gloves if you are handling manure or composted straw. Although the mulch will help to conserve moisture, you will have to apply water during prolonged spells of dry weather.

Canna leaf roller

Although referred to as the Canna Leaf Roller, there are actually two different creatures that go under that name. Cannas are largely free of pests but in the USA plants sometimes fall victim to Calpodes ethlius, the larval state of the Brazilian skipper butterfly, which cuts the leaves and rolls them over to live inside while pupating and eating the leaf. In addition, the Lesser Canna Leaf Roller, Geshna cannalis, a moth larvae, will sew the leaves shut before they can unfurl, by spinning a silk thread around the leaf. The resultant leaf damage can be most distressing to a keen gardener. Gardeners in the rest of the world are free from this particular problem, at present.

Treatment

There are several ways to manage infestations of the Leaf Roller caterpillars. Since there can be three or more generations of the butterfly in a single season, re-treatment should be anticipated.

• If the infestation is light, then affected leaves can be cut off and destroyed or unrolled and cleaned (removing the caterpillar). It is not wise to compost infected leaves, burning is the surest way to avoid re-infection.

• If there is extensive damage, a further measure to consider is Bacillus thuringiensis (BT) which is sold in the USA under various trade names including Dipel and Bio-Worm Killer. Applied as a spray, particular care should be taken to ensure that the underside of the leaves is wet. After applying BT the caterpillars will stop feeding on the leaves, although it may take them a day or two to expire. If BT is applied early enough in the growth of the caterpillars, little damage to the leaves will be evident.

• Another alternative is to apply a systemic insecticide in a foliar spray (i.e. Orthene). When applying systemic insecticides by spraying, careful attention should be given to food and herb crops planted in the immediate vicinity to prevent over-spray or drift of the insecticide.

• A further treatment option to consider is a carbaryl insecticide such as Sevin. Applied as a liquid, Sevin can help manage the infestation of many varieties of insects, unfortunately including beneficial ones.

• Finally, the gardener can try applying a fertilizer with Disulfoton, such as Bayer Advanced Garden 2-in-1 Systemic Rose & Flower Care. This serves to both feed and protect for a period of many weeks.

If using any of the chemical options, then it is recommended that you consult the label for dosage rates and safety precautions. Alternating between the treatment options for management of the Leaf Roller caterpillars can help to spare beneficial insects. Additionally, alternating insecticides also prevents insects from building up resistance to any one particular method.

Canna cultivar Group definitions

The International Code of Nomenclature for Cultivated Plants (ICNCP) regulates the naming of cultivars, cultivar Groups, and graft-chimaerras. Under the ICNCP, a cultivar group is a gathering of cultivars. There may be several reasons to designate a group, for example a group of yellow-flowering cultivars, a group of cultivars with variegated leaves, or a group of cultivars resistant to a particular disease, etc. Also, a cultivar may belong to more than one Group (it may be yellow-flowering, with variegated leaves and resistant to the disease at one and the same time).

At Claines Canna, we are at the stage where we have identified our cultivar groups, proven that they work for us in our 2007 catalogue and web site, but now we need to refine their definitions. Our definitions are shown below, and we would appreciate feedback to allow us to improve on these 'first-cut' group definitions.

Foliage Group
Cultivars, F1 and F2 hybrids, normally with small species-like flowers, but grown principally for their foliage.

Crozy Group
A cultivar group where the flower spikes are arranged close together on the stalk and have narrow to medium petals. There is always space between the staminodes when arranged formally, and the labellum (lip) is smaller than the staminodes, and is often twisted or curled.

The pioneer of this group was Monsieur Pierre Antoine Marie Crozy of Lyons, France, who started breeding Cannas as early as 1862, from stock originally developed by Monsieur Théodore Année, the world's first Canna hybridizer.

They are sometimes referred to as gladiolus flowering cannas, but describing flowers as similar to another genus is not to be encouraged. In the past, they were sometimes called the x generalis L.H. Bailey garden species, but "pretend" garden species have long been replaced by Cultivar Groups in the International Code of Nomenclature for Cultivated Plants.

Italian Group
A cultivar group with large, fragile staminodes. Flowers are arranged somewhat loosely, with wide petals, so wide that there is no space between them, when arranged formally. The labellum (lip) is larger, or at least as large, as the staminodes, unlike the other groups where it is smaller and sometimes curled. The stamen is also much wider than that in the other cultivar groups.

Also, used to be called the "orchid flowering cannas", or ''x orchiodes'' (Liberty Hyde Bailey) garden species, although such "pretend" species are now deprecated in favour of Cultivar Groups. In any event, it is difficult to see the similarity between this group and orchids.

Most of this group obtained its larger sized flowers from the introduction of Canna flaccida in the early 1890's by Carl Ludwig Sprenger in Naples, Italy, followed shortly afterwards by Luther Burbank in California, USA, with the same cross.

Premier Group
This grouping contains cultivars that have a large, circular shape, without significant gaps between the staminodes. Cultivars in this Group are triploids, either produced from a spontaneous or induced mutation during pollination or germination, or from crosses with the Italian group.

Variegated Group
Cultivars with variegated foliage, regardless of what other Group they may belong to.

Conservatory Group
The growing conditions in a Conservatory are quite specialised and do not suit many cultivars, this group have been selected for thriving in this environment, required features being plant vigour, early flowering, foliar appearance, self-cleaning ability and good propagation qualities.

Aquatic Group
Cultivars that thrive as marginal water plants. Characteristically, they will have lance-shaped foliage and long, thin rhizomes.

Miniature Group
Cultivars growing under 0.5m (19") in height, the flowers should be in scale to the rest of the plant.

Agriculture Group
Cultivars grown selectively for agricultural usage, normally for their very large rhizomes and a high starch yield.

Still going after 112 years

Extract from the Garden & Forest Journal of March 20, 1895

The advance of the Canna to its present position as an ornamental plant is one of the marvels of modern horticulture. Unlike the Canna 'Star of 1891', which created a sensation when first introduced, Madame Crozy (see print left) still remains among the indispensable varieties, and a large percentage of recent introductions are seedlings or derivations from this remarkable plant. The yellow border on its petals has given the cue to efforts in the direction of a pure yellow, and this seems to have been the aim of nearly all raisers of new varieties. Improved forms of Madame Crozy have been the result, but the type after all is probably the best. Its moderate height, handsome foliage and short compact spike make it the ideal. Improvements, however, have not been altogether on this type.

In crimsons we have Charles Henderson and Alphonse Bouvier; the former has the larger spike, finer flowers and dwarfer, sturdier habit. Alphonse Bouvier is an abundant bloomer, but is much taller; these two hold a superior rank in their color. Captain Suzzoni is the best yellow of the sorts that now can be called standard. It is not, however, a pure yellow, though the orange-red spots are so faint that for general effect it may be classed as such. Florence Vaughan (see photo right) is an elegant Canna as far as quality of flower and size of spike are concerned, but it is hardly a yellow variety, and should be classed among the spotted varieties. It has the finest yellow ground with scarlet spots of any Canna yet sent out. Captain Suzzoni as a winter bloomer is superior to Florence Vaughan, and in this respect it is equal to Madame Crozy. Taken from the border last autumn it has bloomed continuously all winter, and is now the brightest and best of any. Florence Vaughan received a check, indicating its relation to the older types, which needed some resting. Paul Marquant is a fine satiny salmon, free, and first-class in every way, especially as a winter bloomer. Paul Bruant is a beautiful orange-scarlet, large in flower and one of the first bloomers. This also has proved an admirable greenhouse variety this winter.

Among dark-leaved varieties, J. D. Cabos, with bronzy leaves and salmon flowers; Egandale, with crimson flowers, and the first bloomer of any, and President Carnot, the handsomest in foliage, make a beautiful trio. J. D. Cabos is hardly comparable with either, the glistening salmon-tinted flower contrasting beautifully with the metallic lustre of the foliage. These are the best varieties now in general cultivation.

In novelties every one is inclined to'have favorites of his own. Last summer and autumn I made it a point to note the coming introductions. Eldorado will prove to be one of the finest yellow-flowered sorts this season. It has a large flower, as fine a spike and as good in habit as Madame Crozy, and it is practically yellow. It has some markings, but not enough to detradt from the brilliancy or effectiveness of the flower. Every one who sees it is enthusiastic over it. It is very showy and must become popular. F. R. Pierson will probably be the best scarlet of the season. It bears an ideal flower, brilliant in tone, with only a delicate penciling of pure gold at the centre of the flower, which seems to intensify the scarlet. The petals are very large, round, overlapping each other, reaching near what we must sooner or later obtain in this desirable class of bedding plants --a perfectly round flower.
The Belle of Newport is, so far, the only pure yellow Canna. It is very dwarf, and in this respect promises to be of exceptional value as a bedder. It, however, lacks size, but, no doubt, it will be found valuable for hybridizing with the larger-flowered varieties, and may lead toward a new race of dwarf yellow varieties. The flower of the new socalled Butterfly Canna, Golden Wings, may be described as having a deep yellow ground, heavily blotched with brownish crimson. It will be a fitting companion to Queen Charlotte (see print right), a novelty of German origin-a scarlet, heavily banded with yellow. Both are beautiful and distinct, and especially adapted for effect when used with the darker-leaved varieties in masses. For bedding purposes, and especially for lines, Oriole will probably surpass any of the banded yellows yet introduced. Columbia is a novelty of American origin, and a very distinct break in character. The heads of bloom are most compact, deep cardinal-red, without any shadings; the spikes branch, and the flowers remain on longer than those of any other variety, a most valuable characteristic. Sunshine is much in the way of Paul Marquant, an unusually free bloomer, and bears the largest flower of any Canna I have seen. It has a soft, satiny lustre, and is not excelled by any of its color.

Admiral Avellan I consider an improvement on J. D. Cabos, and this is saying a good deal. It is well worth a trial where a handsome bronzy-leaved variety, which is also a good bloomer, is required. General de Miribel is a good spotted salmon of splendid habit and fine large flowers, and Gladstone is a distinct and pleasing novelty with buff-colored flowers, bordered with yellow. Mrs. Fairman Rogers is a variety in the way of Madame Crozy, which has been honored with a silver medal by the Massachusetts Horticultural Society. The flowers are more heavily banded with yellow than those of Madame Crozy, but its principal value lies in the great size of its branched flower-spike. Its habit is dwarf, and promises to be the best of the Crozy type yet introduced.

Wellesley, Mass. T. D. Hatfield.